Sensitivity and specificity of asymmetric recall on WADA test to predict outcome after temporal lobectomy

Some reports suggest that the intracarotid amobarbital test (IAT) is useful for predicting good seizure outcome after temporal lobectomy. The sensitivity, specificity, and predictive value of the IAT in this condition has not been previously studied. We designed this study to establish the value of memory recall asymmetry on the IAT as a predictor of outcome after temporal lobectomy. We studied memory recall on the IAT in 108 consecutive patients with intractable epilepsy who underwent presurgical evaluation for temporal lobectomy and had at least 1 year follow up after surgery. At a level of 30% asymmetry of recall, specificity for favorable outcome (Engel Class I and II) was 100% (95% confidence interval [CI], 85 to 100), sensitivity 51% (95% CI, 40 to 62), positive predictive value 100% (95% CI, 92 to 100), and negative predictive value 34% (95% CI, 23 to 47). At the same level of asymmetry, specificity for seizure-free outcome (Engel Class I) was 88% (95% CI, 68-95), sensitivity 37% (95% CI, 40 to 64), positive predictive value 87% (95% CI, 71 to 96), and negative predictive value 38% (95% CI, 27 to 50). Asymmetric recall on the IAT is highly specific but not very sensitive in predicting outcome after temporal lobectomy.